Malicious code detecting method applicable to android App

Abstract

The invention discloses a malicious code detection method for Android applications. The server side analyzes the pre-extracted application feature values ​​through a random forest algorithm, thereby establishing a relationship database between applications and malicious codes. Get the data to match and get the test result. Using the advantages of random forest analysis, combined with the idea of ​​application and malicious code classification, a model of application malicious code detection is established based on this method, which can quickly classify and detect existing application malicious code, and can detect new variant malicious codes. The application has certain analysis and detection capabilities, which can meet the ever-changing needs of Android system applications, speed up the detection of malicious applications and codes, and improve the ability to respond to unknown malicious applications.

Description

technical field [0001] The invention belongs to the field of code detection, in particular to a malicious code detection method for Android applications. Background technique [0002] Since the release of the Android system, it has expanded rapidly around the world. According to IDC data, by 2015, Android's market share in mobile operating systems exceeded 80%, and the annual shipments exceeded 1 billion It has become the most widely used smartphone operating system at present. As a result, there are more and more Android applications released. [0003] Due to the openness of the operating system, Android applications have become messy and numerous. A large part of users still think that the security of mobile terminals is not important. There are many third-party mobile markets with chaotic management in the market. Supervision poses a very serious threat to the security of Android terminals. [0004] The openness of the Android operating system lowers its development th...

AI Technical Summary

Problems solved by technology

[0007] In summary, although static detection does not need to run programs, has high code coverage and short detection time, it cannot accurately and comprehensively detect malicious code behavior; although dynamic detection is more comprehensive, it takes a long time and has a negative impact on the operation of the application itself. Great impact; a new Android application malicious code detection method is needed to accurately and quickly detect and classify existing and new malicious code types to meet the ever-changing Android system application development with open features

Images